[OFFICIAL] Bulldozer Reviews Thread - AnandTech Review Posted

Page 22 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

LOL_Wut_Axel

Diamond Member
Mar 26, 2011
4,310
8
81
Yes, it would be better if they went bankrupt, maybe then they'll decide to put more effort into it.

Also, it seems that I am again forced to remind who it may concern that I'm not trying to guilt anyone into buying FXs.

What is that even supposed to mean? The Performance market is much smaller than Mainstream and Essential. Most of AMD's consumer CPU profits come from the Athlon II, Llano, and Bobcat. They do get higher margins on the Performance market, and that's where AMD is hurting the most now. That doesn't mean anything catastrophic for their bottom line, though.
 
Last edited:

RobDickinson

Senior member
Jan 6, 2011
317
4
0
Intel will squeeze AMd everywhere soon, SNB chips spreading across the range, DX11 graphics with the help of the nvidia tie in deal soon too (IVB).
 

qliveur

Diamond Member
Mar 25, 2007
4,090
74
91
If BD had shipped Fall 2009, (prior to Sandy Bridge), people would be falling over themselves to pay $399 for it.
No. It would be getting reamed by Nehalem instead of Sandy Bridge.

Bulldozer is just pathetic.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
http://translate.google.com/transla...2011/amd_bulldozer_fx_prozessoren/index17.php


cray11.jpg
 

BlueBlazer

Senior member
Nov 25, 2008
555
0
76
That C-Ray is on Linux. There's also the x264 encoder (below) and also Avisynth and x264 encoder with AVX support. Seems that the AVX on Bulldozer alone isn't as fast as Sandy Bridge. FMA4 is required to get more performance. :hmm:

i2m2w1.png


Performs the worst in old legacy software like ScienceMark which usually favors AMD. Interestingly the site has Performance index for games, multi-threaded and single-threaded :)
 

Dresdenboy

Golden Member
Jul 28, 2003
1,730
554
136
citavia.blog.de
It looks like AMD's claim at ISSCC 2011 of constant IPC with respect to previous architectures was wildly optimistic.

Time has told.

Time has not told (and writing it bold wouldn't make it true either). Engineers talk about execution of code optimized for a CPU, not for P6, Atom or SB. It even hasn't been tested except for one single app (Cray, als already linked above):
http://ht4u.net/reviews/2011/amd_bulldozer_fx_prozessoren/index17.php

Note: x264 hasn't been recompiled. They used the quickly optimized apps provided by the x264 developer, who just got access to a BD (first via terminal, later as a real chip on his desk) weeks ago.

HT4U even tested the effect of running one thread on a module compared to two threads on a module:
http://ht4u.net/reviews/2011/amd_bu.../index16.php?dummy=&advancedFilter=false&prod[]=AMD+FX-8150+[1+Modul%2C+1+Threads]&prod[]=AMD+FX-8150+[1+Modul%2C+2+Threads]

Instead of wildly exaggerating around people should better start to think. But I guess this is buried deeply in our lower level nervous system.
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
Hardware moves on, software must follow ;)

Or else what?

I can't think of a single software program that was doomed if it didn't follow hardware. Classic case of false logic "if you build it, they will come". The customer of hardware is software, hardware exists to serve the customer. Not the other way around.

If it were the other way around then we'd still have DEC Alphas, Itanium chips, PA-RISC, MIPS, Cray, Transmeta, Via, etc. And yet most of these have relegated themselves to niche markets, or bankrupted themselves in time, for failing to serve their customers.

AMD is following a well-trod road. Calling this duck anything but a duck is just delusion at its finest.
 

AtenRa

Lifer
Feb 2, 2009
14,003
3,362
136
Or else what?

I can't think of a single software program that was doomed if it didn't follow hardware. Classic case of false logic "if you build it, they will come". The customer of hardware is software, hardware exists to serve the customer. Not the other way around.

If it were the other way around then we'd still have DEC Alphas, Itanium chips, PA-RISC, MIPS, Cray, Transmeta, Via, etc. And yet most of these have relegated themselves to niche markets, or bankrupted themselves in time, for failing to serve their customers.

AMD is following a well-trod road. Calling this duck anything but a duck is just delusion at its finest.

I believe you are wrong here,

If you dont have the hardware, you dont create the software.

Developers will not code in AVX if Intel/AMD will not have AVX capable CPUs.

Game developers will not create DX-11 games if AMD/NVIDIA will not have DX-11 Hardware(GPUs)

Hardware leads, Software follows ;)
 

Idontcare

Elite Member
Oct 10, 1999
21,110
64
91
I believe you are wrong here,

If you dont have the hardware, you dont create the software.

Developers will not code in AVX if Intel/AMD will not have AVX capable CPUs.

Game developers will not create DX-11 games if AMD/NVIDIA will not have DX-11 Hardware(GPUs)

Hardware leads, Software follows ;)

Nope. Now you are conflating the difference between "if you don't build it, then they can't come" versus "if you build it, they will come".

Hardware does not lead, it merely creates opportunity. AVX apps do not exist just because AVX was created. Look at 3DNow. AVX apps exist because AVX exists and AVX was something software could benefit from taking advantage of.

There is a critical difference, and you are missing that in this blind pursuit of "because they built bulldozer, the software will follow". Software will only follow bulldozer if there is a benefit to doing that versus the alternatives.

SSE did not take off in software because SSE was the future...SSE took off because it had intrinsic advantages over continued use of x87 instructions.

Last I checked Windows was no longer being developed for Itanium, or MIPS. There is a good reason for this. If Bulldozer diverges from mainstream too far, it may well end up where Itanium is, living on non-Windows OS for server apps. That is not software following hardware.

I agree that without hardware there will be no software, but one is not leading the other as a General leads troops into battle. Hardware merely creates opportunity for software, but software will not blindly follow hardware just because it exists. (run many CUDA apps lately? how's your Itanium desktop doing? 3DNow?)

If you look to AMD's past, the strength of the K7 was that it ran apps of the day with aplomb, same with the K8. When Prescott showed up, was its future secured by program recompiles? Why was C2D the success it turned out to be? Was it because software followed hardware?
 

Meghan54

Lifer
Oct 18, 2009
11,684
5,228
136
I believe you are wrong here,

If you dont have the hardware, you dont create the software.

Developers will not code in AVX if Intel/AMD will not have AVX capable CPUs.

Game developers will not create DX-11 games if AMD/NVIDIA will not have DX-11 Hardware(GPUs)

Hardware leads, Software follows ;)


I see you're thinking about niches and not the overall picture of software development. Outside of computer gaming, a very small niche, software pushes hardware. Consider the scientific end of computing. Software outpaces the hardware constantly which is why the supercomputers are always being upgraded....the software evolves faster than the hardware and struggles. This is seen consistently....weather prediction, engineering, and darned near any other heavy scientific use. Hardware lags behind what the software writers are coding.

And as for games, Crysis is but one example of a game that pushed hardware to its limits when it was released, and hardware still hasn't caught up. As for DX11, in the broadest terms, the DX11 code (software) was written well before the first piece of hardware was available to take advantage of it. And the initial DX11 games were well down their development pipelines before the first DX11 capable gpu was commercially released.
 

Abwx

Lifer
Apr 2, 2011
11,885
4,873
136
Or else what?

I can't think of a single software program that was doomed if it didn't follow hardware. Classic case of false logic "if you build it, they will come". The customer of hardware is software, hardware exists to serve the customer. Not the other way around.

So no one did use SSE2 and beyond , since it wasnt the customer
that did ask for those new instructions but the main CPUs manufacturer...
SSE did not take off in software because SSE was the future...SSE took off because it had intrinsic advantages over continued use of x87 instructions.

SSE2 was implemented because intel has no other easy mean
to match K7 s superior FPUs , that s the main point.
 
Last edited: